Detecting low-voltage arc fault based on lifting multiwavelet

Arc fault is one of the prime reasons causing electrical fire accidents. It is difficult to detect the arc faults´ features by traditional circuit protections. An algorithm using lifting multiwavelet transform for arc fault detecting was presented in this paper. First, experiments were conducted according to UL1699 standard and data were collected by LabView. The high frequency band information of arc faults could be detected by lifting multiwavelet transform, and then arc feature parameters of the wavelet coefficients were calculated for arc detecting. The experiments using typical loads showed that our method performs better for arc fault circuit interrupter when compared to the traditional wavelet.